One approach to treating cocaine addiction is to vaccinate against cocaine, preventing it from reaching the brain. The challenge is that cocaine, like most small molecules, is a poor immunogen. Based on our experience with adenovirus (Ad) gene transfer vectors in animals and humans, and the recognition that Ad is a potent adjuvant that activates the immune system, we hypothesized that cocaine analogs linked to Ad capsid proteins would elicit high level, high affinity cocaine specific antibodies sufficient to trat cocaine addiction. Our 1st approach was to link the cocaine analogs GNC or GNE to a disrupted E1-E3- serotype 5 Ad (dAd5). The dAd5GNC vaccine elicited high affinity (Kd 45 nM) IgG anti-cocaine titers in mice, and vaccinated mice no longer responded with hyperactive behavior following repetitive intravenous doses of 50 ?g cocaine. dAd5GNE evoked higher titers than dAd5GNC, and dAd5GNE-vaccinated rats exhibited suppressed cocaine reward, no extinction burst of activity seen in non-vaccinated rats, and did not reinstate cocaine seeking following a cocaine prime. The focus of the 3 aims of this proposal is to develop 2nd generation anti-cocaine vaccines that build on the success of dAd5GNC and dAd5GNE, enabling lower doses to generate higher titer, higher affinity anti-cocaine antibodies that abrogate the activity and self-administration phenotypes associated with cocaine administration. Aim 1. The capsid hexon and fiber are the most immunogenic Ad proteins. We hypothesize that the cocaine analog GNE coupled directly to purified hexon and/or fiber will generate a more potent vaccine than GNE coupled to the entire disrupted Ad. Further, the hexon and fiber sequences will be genetically modified to include additional lysine residues, increasing covalent attachment sites for GNE. Aim 2. Immunity against Ad capsid proteins limit efficacy of Ad vaccines. Since different Ad serotypes stimulate immunity differently, we hypothesize that repeated alternating administration of GNE coupled to disrupted Ad5 and sAd36 (a highly immunogenic nonhuman primate Ad) or alternating administration of GNE coupled to purified hexons/fibers of each serotype, will evoke higher titers than with a single serotype. Aim 3. We have also developed a persistent passive anti-cocaine immunity strategy using an adeno-associated virus serotype rh.10 coding for an anti-cocaine monoclonal antibody (AAVrh.10antiCoc) that generates persistent anti-cocaine antibodies sufficient to suppress cocaine induced hyperactivity in mice. We hypothesize that AAVrh.10antiCoc either alone, or together with the best active Ad vaccines from aims 1 and 2, will provide highly effective anti-cocaine protection to the CNS. An AAV coding for anti-cocaine will also be administered to the nose to determine if high titer anti-cocaine antibodies at the site of entry can aid as a 1st anti-cocaine defense.